Out of curiosity, I set out to do some hires profiling of a stock
kernel, one with Richards patch, and one with Rik's patch. Along
the way, I ran into something I don't understand. (The following
isn't quite the way I intended to get real results, only a demo of
what I came across that I found to be strange.) If I call the script,
it doesn't work, but if I take it it does... anyone know why?
-Mike
Script started on Thu Oct 1 08:41:58 1998
[root]:# cat testo
#!/bin/sh
rm -f it_ran times.larry times.richard
echo 'START lat_ctx' >> it_ran
echo > /proc/profile
for i in 0 1 2
do
echo -n $i >> it_ran
./lat_ctx -s 32 32 >>times.larry 2>&1
done
echo >> it_ran
echo 'END lat_ctx' >> it_ran
readprofile | egrep 'schedule|__switch_to' >> times.larry
echo 'START time-schedule' >> it_ran
echo > /proc/profile
for i in 0 1 2
do
echo -n $i >> it_ran
./time-schedule -pipe 10 >>times.richard 2>&1
done
echo >> it_ran
echo 'END time-schedule' >> it_ran
readprofile | egrep 'schedule|__switch_to' >> times.richard
[root]:#
[root]:#
[root]:#
[root]:# testo
Terminated
[root]:# cat it_ran
START lat_ctx
012
END lat_ctx
START time-schedule
0[root]:#
[root]:# . testo
Terminated
Terminated
Terminated
[root]:# cat it_ran
START lat_ctx
012
END lat_ctx
START time-schedule
012
END time-schedule
[root]:# exit
exit
Script done on Thu Oct 1 08:43:08 1998
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/